 Global Adaptive Transformer, GAT, is a new domain adaptation method which uses parallel convolution to capture temporal and spatial features in EG signals. It then employs an adapter to transfer source features to the target domain while simultaneously reducing the marginal distribution discrepancy between the source and target domains. This is achieved through an adaptive center loss and an explicit discriminator. The proposed method outperformed existing methods on two widely used EG datasets, demonstrating its superior performance. This article was authored by Yonghao Song, Qingqing Zheng, Qiong Wang, and others.